Mixed-halide perovskite spectra generators

This application claims the benefit of priority of Provisional U.S. Patent Application Ser. No. 63/660,788, filed 17 Jun. 2024, which is incorporated herein by reference in its entirety for all purposes.

The present disclosure generally relates to systems and methods for solar simulation, and in particular to systems and methods for solar simulation with a mixed-halide perovskite spectra generator.

Solar simulators are widely used in scientific, industrial, and commercial settings to replicate natural sunlight under controlled conditions. These devices emit electromagnetic radiation across a range of wavelengths that approximate the spectral power distribution of sunlight at the Earth's surface, often modeled on standardized solar spectra such as Air Mass 1.5 Global (AM1.5G). Solar simulators enable consistent and repeatable testing of photovoltaic devices, optical materials, and chemical processes that are influenced by solar radiation. Common applications include the evaluation of solar cell efficiency, accelerated material aging studies, photochemical reaction testing, and calibration of light-sensitive instrumentation.

Conventional solar simulators typically employ xenon arc lamps, metal halide lamps, or, more recently, light-emitting diodes (LEDs) to generate a light source with an appropriate spectral match, intensity uniformity, and temporal stability. The performance of a solar simulator is often classified in accordance with standards such as ASTM E927 or IEC 60904-9, which define criteria for spectral accuracy, irradiance uniformity, and stability over time. Despite ongoing improvements, existing simulators may suffer from issues such as high-power consumption, limited tunability, spectral mismatches, or non-uniform irradiance across the test plane. These limitations can reduce measurement accuracy and hinder reliable testing of advanced solar technologies.

The following is a non-exhaustive listing of some aspects of the present techniques. These and other aspects are described in the following disclosure.

In some aspects, the techniques described herein relate to a system, including: a first emitter that includes: a first mixed halide perovskite film that includes a first type mixed halide perovskite material that possesses a first usable photoemission wavelength range; and a first excitation device that is positioned adjacent to the first mixed halide perovskite film based on an excitation mode of the first excitation device and the first excitation device is configured to: tune, based on a first instruction, the first mixed halide perovskite film to emit a first photoemission peak within the first usable photoemission wavelength range; and tune, based on a second instruction, the first mixed halide perovskite film to emit a second photoemission peak within the first usable photoemission wavelength range, wherein the second photoemission peak has at least one of a different width or center within the first usable photoemission wavelength range than the first photoemission peak.

In some aspects, the techniques described herein relate to a system, including: a controller; a driver that is coupled to the controller, that receives instructions from the controller, and that is configured to control voltage characteristics of a signal; and an electroluminescent emitter device that is coupled to an output of the driver to receive the signal and that includes: a light source; and a mixed halide perovskite film positioned adjacent to the light source such that photoemission emitted from the light source excites the mixed halide perovskite film to generate a photoemission peak of a usable photoemission wavelength range of a type of mixed halide perovskite material used in the mixed halide perovskite film.

In some aspects, the techniques described herein relate to a process, including any of the above-mentioned processes performed by the systems.

Some aspects include a tangible, non-transitory, machine-readable medium storing instructions that when executed by a data processing apparatus cause the data processing apparatus to perform operations including the above-mentioned processes performed by the systems.

While the present techniques are susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. The drawings may not be to scale. It should be understood, however, that the drawings and detailed description thereto are not intended to limit the present techniques to the particular form disclosed, but to the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the present techniques as defined by the appended claims.

To mitigate the problems described herein, the inventors had to both invent solutions and, in some cases just as importantly, recognize problems overlooked (or not yet foreseen) by others in the fields of solar simulators and mixed-halide perovskites. Indeed, the inventors wish to emphasize the difficulty of recognizing those problems that are nascent and will become much more apparent in the future should trends in industry continue as the inventors expect. Further, because multiple problems are addressed, it should be understood that some embodiments are problem-specific, and not all embodiments address every problem with traditional systems described herein or provide every benefit described herein. That said, improvements that solve various permutations of these problems are described below.

Modern lighting technologies are capable of producing sufficient luminous output to illuminate indoor environments and enable navigation in low-light conditions. However, significant limitations remain regarding the spectral quality of artificial light, particularly in efforts to replicate the solar spectrum. Sunlight that reaches the Earth's surface predominantly falls within the 300 to 1600 nm wavelength range and is characterized by numerous atmospheric absorption features that create pronounced spectral valleys. Achieving a Class A+solar spectral match—the highest classification currently recognized by the International Electrotechnical Commission (IEC)—requires compliance with integrated spectral intensity limits across six defined bands: 300-470 nm, 470-561 nm, 561-657 nm, 675-772 nm, 772-919 nm, and 919-1200 nm. Within these bands, the maximum permissible error in integrated spectral intensity is 12.5%. The IEC acknowledges that even advanced solar simulators may be unable to substantially exceed the performance required for this classification. Challenges in spectral matching are also observed in narrower spectral subsets, which are often targeted in applications that do not require full-spectrum solar replication. The techniques and systems described herein are not limited to addressing only the aforementioned challenges, but may also be applied to solve additional problems or improve performance in specialized contexts as will be apparent to those of ordinary skill in the art based on the present disclosure.

Some embodiments of the present disclosure are expected to go beyond currently achievable spectral fidelity, so a higher standard for calculating spectral error is also helpful to quantify the relative differences exhibited by some embodiments discussed below. It may be helpful to use a quantitative metric which accounts for the error in the irradiance within 1 nm spectral bands between the target spectrum and the achieved spectrum. An intuitive metric would be the average or root-mean-square relative error of each band. It should be noted however, that due to deep atmospheric absorption features in sunlight, the relative spectral error in some bands would be exceptionally large and this metric would more heavily emphasize depleted regions of the spectrum than bright portions.

Therefore, it may be helpful to use a scheme which weights by the absolute spectral error in each band but normalizes to the integrated spectrum to express the result as a percent error. An expression for this error is then:

In equation (1), Iis the target spectrum over the wavelength range [λ,λ](where λ refers to wavelength of light) and Si is the spectrum produced by the solar simulator over the same wavelength range. Both Si and I; are expressed as irradiance with units of

One type of solar simulator (e.g., a light source that emits light approximating solar spectrum) is fluorescent lamps. A Xenon lamp may produce emission throughout the majority of the visible range and when operated at a high temperature, the radiative emission picks up the peak wavelength and blackbody emission lineshape similar to the sun. However, major peaks and valleys are often nonexistent or appear where they should not be. Light emitting diodes (LEDs) typically have a characteristic emission color, however white LEDs are often commercially made by starting with a blue LED and down converting a substantial portion of the emitted light through fluorophores broadening the emission towards something approaching a blackbody spectrum. A constructive approach has also been made to produce solar simulators by juxtaposing multiple LEDs with different emission spectra.illustrates a graphof a standard illumination spectrum (e.g., AM1.5) versus prior art artificial light sources such as a Xenon lamp that has a 54% error, “white” LED with a 75.5% error, an LED-constructed spectrum with 27% error. The errors are reported for the 430-1350 nm range.

Embodiments of the present disclosure produce a close match to true sunlight (or other target spectra). Some embodiments utilize a single class (or more than one class) of color-tunable materials which in some cases may have highly advantageous properties for this function: organic metallic mixed-halide perovskites. In particular, these materials have high luminescent quantum yields which are further enhanced by the “carrier funneling” effect and have an emission color and which can be controlled via the operating condition of the material as described in the “photosegregation” effect and “photomixing” effect (also referred to “photoremixing” effect herein), in some embodiments. Furthermore, emission peak widths are minimized (or reduced) by the “bandgap homogenization” effect which significantly aides spectrum fitting.

These materials may have composition (e.g., chemical formula of a crystallographic structure in the form of a perovskite) ABX3 (where A is an alkali or pseudoalkali: A=methylammonium (MA+), formamidinium (FA+), Guanadinium (GA+), Cesium (Cs+), Rubidium (Rb+) or an alloy thereof: B=Lead (Pb2+) or Tin (Sn2+), and X is a halide or pseudohalide: X=Fluoride (F−), Chloride (Cl−), Bromide (Br−), Iodide (I−), Formate (Fo−), or Thiocyanate (TC−) or an alloy thereof). The emission wavelength of single emitters can, in some embodiments, be tuned both by the halide alloy of the material as well as the carrier injection level. Multiple individual emitters may be excited within the same film with each emitter producing a separate color.

illustrates a graphof a constructed spectrum formed within an example model mixed halide perovskite material meant to capture the breadth of achievable bandgaps between 0.8 and 3.2 eV and interrelatedness of carrier injection level, emission wavelength, and emission peak width. Namely, in this example, the emission peak wavelength increases with carrier injection level while the peak width remains small from the photosegregation phenomenon. However, at high carrier injection levels, in this example, photomixing begins which reduces the emission peak wavelength while first increasing and then substantially decreasing the peak width. While this model oversimplifies the construction as a single material, it also underestimates the capability to reduce spectral error through additional sharply-peaked emitters that would come from using multiple perovskite compositions. These sharply-peaked emitters would substantially fill in poorly-fit regions of the spectrum.

In the specific modeled constructed spectrum of the graphof, 127 mixed-halide emitters were used with an emission peak range of 385 nm to 1550 nm optimized to approximate the solar spectrum within the 430 to 1350 nm range. The spectral match shown inhas an error of 3.8% over the 430 to 1350 nm range as calculated by equation 1.

illustrates a graphof the percent error of calculated spectra of some embodiments versus the number of emitters. The graphillustrates the spectral error relative to AM1.5G Standard illumination spectrum versus number of emitters assuming an emission peak range of 385 nm to 1550 nm optimized to best match the solar spectrum within the 430 to 1350 nm range. The optimization algorithm started with a single emitter and performed a gradient descent to find a local minimum in the carrier injection level. Each subsequent emitter was added one at a time starting with a random injection level and all emitters were optimized by gradient descent. As this is a local minimum finding routine, the solution is not necessarily a global minimum.

Various ranges may be targeted depending on the use case. Examples include: tandem solar cells with spectral fidelity in the 350-1140 nm range, spectrum matching within the human visual acuity range (400-700 nm), chlorophyll matching (400 to 500 nm and 650 to 680 nm), “soft” illumination (550-1000 nm), water-penetrating lighting (350-450 nm), biochemical (550-850 nm), or other ranges that would be apparent to one of skill in the art in possession of the present disclosure. Within the natural sciences as a scientific tool a helpful scenario is a high-fidelity light source with quantifiable spectral fidelity over as wide a range as possible (200-1800 nm).

Furthermore, as the solar spectrum changes throughout the day/year as well as with the latitude and altitude, some embodiments may produce multiple spectra at different times, e.g., responsive to control signals from a controller having a clock and executing a program by which such updates are applied, e.g., with a formula or lookup table. It is expected that this function may be especially useful for biological systems, which are known to synchronize their internal clock in response to changes in the illumination spectrum. Indeed, while it is known that a bluer spectrum is stimulating while a redder spectrum promotes sleep, devices that targeted this effect have generally been limited in the accuracy of their spectrum. For instance, some embodiments may apply the spectra and modulations described in Patterson et. Al, A Color Vision Circuit for Non-Image-Forming Vision in the Primate Retina, in Cell, VOLUME 30, ISSUE 7, P1269-1274.E2, Apr. 6, 2020, available at https://doi.org/10.1016/j.cub.2020.01.040, the contents of which are hereby incorporated by reference, to obtain the biological effects described therein. As a broad classification, study and control of photochemical processes in biological systems are expected to greatly benefit from this light source.

The illumination spectrum may, in some cases, be further modified by environmental factors such as cloud coverage as well as surface and ground albedo. The right color spectrum may form an effective trap for pest insects and animals.

In some cases, no single mixed-halide material may fully cover the targeted range, but the range may be covered with multiple material compositions, as discussed below. The emission spectrum produced through the combination of all emitters in this approach is significantly closer to the solar spectrum than that produced via adding many disparate LEDs together. The disparate LEDs may have differing emission peak widths and non-uniform and uncontrolled spacing in emission peak wavelength, which makes creating an optimal choice of LEDs difficult. Furthermore, the improvement that can be realized from adding an additional LED to an existing solution is marginal because of overlap between the spectra of the emitters. In contrast (which is not to imply that the present techniques may not also be used with LEDs), the emission peak of the single emitters within the perovskite film is tunable. As additional emitters are added other existing emitters can have their peak wavelength adjusted. Furthermore, since the emitters are in close proximity, the light produced can be treated as a point source making this light source focusable. Being a point source also minimizes the number of discrete emitters that need to be controlled as the disparate LED case requires many LEDs of each color in order to balance out spatial nonuniformities.

illustrates an example architectureof a mixed-halide perovskite based arbitrary spectrum generator that may include an arrayof individually addressable pixels-that inject carriers into mixed-halide perovskite emitters-. In some embodiments, several emitters are within each perovskite composition (e.g., at each addressable pixel). Each emitter-, in some embodiments, produces different emission peak wavelength and peak width based on the perovskite composition and the applied injection level for that emitter. In some cases, each pixel-may have a respective instance of the same set of emitters, or different pixels may have different emitters configured to emit at different wavelengths than those in other pixels. In some embodiments, the above architecture may be used for a factory-fixed spectrum and in some embodiments, the above architecture may be used to produce multiple discrete spectra with the same pixels. Finally, in some embodiments, the above architecturemay be used to produce arbitrary spectra. Depending on use case, there are many ways to drive the emitters-: dedicated analog output for each emitter, a multiplexing scheme setting latch circuits for each channel, factory-set impedance control, time-split PWM (pulse-width modulation) driving.

Embodiments of the present disclosure may have many use cases. For example, the mixed-halide perovskite spectra generator may provide for factory-fixed light sources (direct sunlight: AM1.5G, AM1, AM0, AM2, dusk/dawn: AM3, AM4, AM7), daylight simulators (AM4 at dusk and dawn, AM3, AM2, AM1, etc. in the middle), arbitrary spectrum generators-advanced scientific tool, reprogrammable on the fly-true color displays (like an LED display but better spectral fidelity), (e.g., in televisions, monitors, watch faces, virtual reality or augmented reality displays, and the like), plant grow lights, pest control, forensic contrast highlighting of certain substances, and other use cases that would be apparent to one of skill in the art in possession of the present disclosure.

Another use of aspects of the present disclosure involves hyperspectral imaging. Existing hyperspectral imaging systems commonly utilize broad-spectrum (white) illumination sources in combination with electrically tunable optical filters to isolate individual wavelength bands. These tunable filters are often complex and cost-prohibitive and operate by sequentially rejecting the majority of incident spectral content, resulting in significant inefficiencies. In contrast, aspects of the present disclosure enable direct tuning of the excitation light wavelength, eliminating the need for downstream filtering and thereby increasing signal strength while reducing system cost and complexity. Additionally, existing tunable filters typically exhibit broad transmission bandwidths, whereas the disclosed light source architecture is capable of producing emission peaks with narrower spectral widths, offering improved spectral resolution for hyperspectral imaging applications.

Another advantageous use case for aspects of the present disclosure is in fiber-optic-based structural health monitoring systems. Such systems are employed in critical infrastructure, including dams, pipelines, subsea cables, and large buildings, where they enable distributed sensing of mechanical parameters such as strain along the length of an optical fiber. The implementation of a color-tunable, coherent light source within these systems would provide substantial performance enhancements. In particular, wavelength tuning of the excitation light allows for spatial discrimination of localized responses along the fiber, thereby enabling more precise identification of structural anomalies or stress concentrations. The disclosed light source offers tunability and coherence characteristics that are well-suited to improve resolution and diagnostic capability in such distributed fiber sensing networks.

A mixed-halide perovskite spectra generator may include the perovskite in various form factors. For example, the perovskite crystal structure may be formed in place by depositing precursors and converting to form a thin film. The film thickness may be of 30-3000 nm. The film may be formed by spin-coating, formed by blade coating, formed by slot-die coating, formed by thermal evaporation, formed by sputtering, converted by thermal annealing, converted by microwave annealing, converted by antisolvent method, deposited in a one-step process whereby a crystal is precipitated from a solvated solution containing alkali, lead, and halide ions, deposited in two or more coats containing lead halide and an alkalihalide or pseudoalkali halide, or other fabrication processes that would be apparent to one of skill in the art. In the fabrication process where two or more coats are deposited, both coats may be made using solution-processing techniques or using dry processing techniques, or when at least one coat is made using a dry process and at least one coat is made using a wet process. In some embodiments, a crystal of perovskite structure is synthesized and then mounted in a device. In some embodiments, multiple crystals of perovskite are synthesized and then deposited as a colloidal film. In some embodiments, the crystals may be nanocrystals (<30 nm in all dimensions), the crystals may be nanowires (<30 nm in two dimensions), the crystals may be nanoplatelets (<30 nm in one dimension, or those crystals may be microplatelets (>1 μm in two dimensions).

In some embodiments, device architectures include when carriers are injected into the perovskite through electrical connections. For example, the device architecture may be NIP (N-type layer, intrinsic layer, P-type layer)—common perovskite device architectures featuring an n-type transition metal oxide (SnO2 or TiO2) electron transport layer grown on a transparent conducting electrode (ITO or FTO), followed by an intrinsic perovskite photoactive layer, and a p-type organic (spiro-OMETAD, PTAA) or inorganic (NiOx) hole transport layer. In other examples, the device architecture may be PIN-common perovskite device architecture featuring a p-type organic (PTAA, SAMs) or inorganic (NiOx) hole transport layer grown on a transparent conducting electrode (ITO or FTO), followed by an intrinsic perovskite photoactive layer, and a n-type organic (C60, PCBM, BCP) or inorganic (SnO2) hole transport layer. Another example device architecture may include architectures when the carriers are injected into the perovskite through photoabsorption. The perovskite may be illuminated as a fluorophore-a device architecture where an underlying hard semiconductor (III-V or II-VI) provides the initial light and pixel addressing, but a perovskite down converts the light and determines the final emitted color. This may be used for QLEDs.

In various embodiments of the present disclosure various perovskite compositions may be used in the film(s). For example, these perovskite compositions may include MAPb(IBr)—The entire range of [0.2<x<0.9] is usable, where MA refers to methylammonium. In some embodiments, a range of [0.15<x<0.95] may be contemplated. Below x=0.2, there may be insufficient driving force for photosegregation. Above x=0.9, the kinetics of I-rich domain nucleation may become extremely slow and unusable. Another perovskite composition may include MAPb(BrCl), where MA refers to methylammonium. Where a range [0.2<x<0.9] is usable. Another perovskite composition may include MAPb(IBrCl), where MA refers to methylammonium. The range of iodine concentrations which result in photosegregation in cubic perovskites is extended because of larger bandgap differences between I and Cl. However, it may be the case that the extremes (I and Cl) are immiscible without Br to bridge the size range. Some embodiments may use the ranges of [0.15<x<0.95] and [0.15<y<0.95]. In various embodiments, another perovskite composition may include FAPb(IBrCl), where FA refers to formamidinium. This results in a slightly lower bandgap than MAPb(IBrCl)due to FA being a little bit larger than MA. However, pure FA is rare because it is a bit too large for the structure it inhabits and favors a hexagonal phase over the necessary cubic phase. It may be the case that cubic FAPbI3 can be kinetically stable, some embodiments may include pure FA. In various embodiments, another perovskite composition may include CsPb(IBrCl)-pure Cs is a stable means to increase the bandgap a little bit. A pure inorganic material has certain advantages for stability and processing and works well in nanocrystals. Cs is a little bit too small for iodine and CsPbI3 is orthorhombic, but Cs works well for bromine, I/Br mixtures, Cl and Br/Cl mixtures. It does tend to have reduced photocarrier lifetime, however compared to perovskites containing organic cations. In yet another embodiment, the perovskite composition may include FACS: Pb(IBrCl), where FA refers to formamidinium-FACsPbI3 may be used for single-junction perovskite solar cells. Example compositions include: FACsPb(IBrCl), FACsPb(IBrCl), FACsPb(IBrCl). FACsPb(IBrCl), FACsPb(IBrCl), FACsPb(IBrCl)In yet another embodiment, the perovskite composition may include CsRbPb(IBrCl)—As an all-inorganic option that may be used in some embodiments. In yet another embodiment, the perovskite composition may include FAMA(CsuPb(IBrCl)where MA refers to methylammonium and FA refers to formamidinium—may be used in some embodiments. In yet another embodiment, the perovskite composition may include FAMARbCsPb(IBrCl)where MA refers to methylammonium and FA refers to formamidinium. In yet another embodiment, the perovskite composition may include GAPb(IBrCl)where GA refers to guanidinium-GA is a bit larger still than FA and may be useful for lower bandgaps or in conjunction with pseudohalides. In yet other embodiments, tin-based perovskites-gives a significantly lower bandgap than lead and may be used in some embodiments. In yet other embodiments, the perovskite composition may include CsFAPb(BrCl), CsPb(BrCl), MASn(BrCl), MAPb(I(BrCl)), FACsPb(IBr), FASn(IBr), or MASn(IBr).

In some use cases, other compositions may be used. For example, a composition of perovskite may include Ruddleson-Popper (RDP)—2D perovskites containing benzylammonium, butylammonium or phenethylammonium,—A two-dimensional crystal structure of perovskite composition featuring large cations with a positive ammonium group on one side and an aliphatic tail on the other. These can form a 2D/3D hybrid where the 3D part is quantum confined giving an increased bandgap. These materials may be used either as the primary absorber, or as a capping or interlayer between the perovskite and electrodes. They usually have valence band alignment with 3D perovskites. Another perovskite composition may include Dion-Jacobson 2D perovskites-A two-dimensional crystal structure of perovskite composition featuring large cations with two positive ammonium groups on either end of an organic chain. These can form a 2D/3D hybrid where the 3D part is quantum confined giving an increased bandgap. These materials may be used either as the primary absorber, or as a capping or interlayer between the perovskite and electrodes.

Mixed-halide perovskite materials are ionically-active heterogeneous semiconductors. When excited to form electron-hole pairs, the halides within these materials redistribute altering their electronic properties. This minimizes (or reduces) the free energy of photocarriers in accordance with a thermodynamic bandgap model. This behavior is analogous to the erosion of the geological landscape by water. Water not only travels to the lowest accessible point: it also carves deep grooves through the landscape. The alterations made to the landscape affect the way that future water travels, and these changes accumulate over time giving rise to rivers, lakes, and seas. However, unlike erosion, the changes within the perovskite are typically fully reversible.

It is noted that some claim to have suppressed instabilities related to halide segregation. In turn, this implies that the entire spectrally-accessible range of mixed halide perovskites can be used to produce spectrally stable homogenous semiconductors. However, the routes taken have either (1) resulted in other undesirable characteristics such as poor charge carrier lifetime or diffusion rate, (2) slowed but not stopped the effect where instabilities still occur over longer timescales, or (3) produced stability only within a narrow range of emission wavelengths such as the well-known 0 to roughly 20% stability region for the high bandgap halide in the alloy. Some embodiments therefore do not require a spectrally stable mixed halide perovskite, rather a consistent heterogeneous mixed halide perovskite may be used in some embodiments to produce the desired emission spectrum.

The thermodynamic bandgap model can be summarized in terms of 3 behaviors. (1) Photocarriers funnel to the lowest accessible bandgap. If there are sufficient photocarriers to fill the lowest bandgap, then they begin filling the next lowest, etc. The photoemission of the material is produced by the combination of all photocarriers and is nearly equivalent to the bandgap at the locale of the photocarrier. (2) Photocarriers burrow and will produce a region with a lower bandgap than its surroundings via expulsion of halides responsible for a high bandgap and uptake of halides responsible for a low bandgap. (3) Areas that contain photocarriers generally homogenize such that they approach a single consistent bandgap.

The combination of (1) and (2) results in the so-called “photosegregation” phenomenon whereby mixed-halide alloys which are excited will over time produce emission that is red-shifted compared to the initial emission. When (1) and (3) are combined it results in a phenomena termed photomixing whereby the halide composition can be more evenly mixed than the random stochastic process responsible for mixing in the dark. These two phenomena and the other behavior allow for different emission wavelengths/energies depending on the excitation density while at the same time minimizing the emission peak width through the homogenization effect. Notably, both photosegregation and photomixing occur on relatively short timescales of seconds to minutes allowing for their utilization as changing materials in functional devices. In turn, this behavior can be utilized and tuned to produce a light source with useful photoemission.

illustrates graphsandof kinetic Monte Carlo simulations of first photosegregation and then photomixing within the Thermodynamic Bandgap Model. A result of these simulations is the aforementioned significant sharpening of the electronic structure during photomixing due to the propensity for creating a uniformly mixed state. Graphillustrates simulations of photoemission spectrum and graphillustrates simulations of photoabsorption spectrum of MAPbI1.5Br1.5. Plotillustrates an initial state of the mixed halide perovskite. Simulations were first photosegregated at a low injection level (plot) and then photomixed at a high injection level (plot). These are compared to simulations of remixing in the dark (plot) to distinguish photomixing and dark remixing. It should be noted that these simulations represent a small sliver of the overall phenomenon meant to understand the underlying physics. The MAPb(I1-xBrx)3 was studied starting from MAPbI1.5Br1.5 which subsequently segregates to MAPbI2.4Br0.6 covering a range of 1.65 to 1.83 eV. However, this single alloy can be utilized from MAPbI2.4Br0.6 to MAPbI0.3Br2.7 giving an energy range of 1.65 to 2.18 eV. This same behavior could be used for MAPbBr2.4C10.6 to MAPbBr0.3C12.7 giving a range of 2.35 to 2.88 eV.

As discussed above is that photosegregation is reversible. When photosegregated films are kept in the dark over a period of hours, original mixed-halide absorption and emission energies/spectra recover. X-ray diffraction measurements confirm this, showing restoration of I/Bralloying. Dark remixing is attributed to entropically-driven remixing of photosegregated anions. As such, mixed-halide perovskite thin films are quick to photosegregate or red shift but slow to remix in obtain shorter wavelengths (e.g., slow blue shift). Unfortunately, hour long, entropically-driven remixing timescales are impractical for applications to quickly utilize the usable wavelength range of a material such as to obtain a short wavelength photoemission from a shorter wavelength photoemission of photosegregated mixed-halide perovskite.

The inventor of the present disclosure has now discovered that persistent photoremixing can be induced in photosegregated, mixed-halide perovskite thin films in faster time scales using several mechanisms. It has been discovered that while using gradually increasing low-intensity light sources (e.g., 1-50 uJ/cm) shifts the wavelengths longer and photosegregates the mixed-halide perovskite thin films. Increasing the intensity of the light source by several factors of 10 can achieve photoremixing such that providing high-intensity light has the same effect as darkening on the mixed halide perovskite but the high-intensity light can photoremix the mixed-halide perovskite thin films on much faster time scales (e.g., a matter of seconds or minutes versus hours). One way of using high-intensity light is to use a continuous wave. Using a light source with a continuous wave requires an intensity of 100-10,000 kW/cmto obtain photoremixing. However, intensity-based control over the emission wavelength is difficult to achieve. This is because of the huge range of intensities involved (uW/cmto kW/cm). It is difficult to obtain intensities from LEDs that are greater than 1000 W/cm.

Another intensity-based method to achieve photoremixing includes pulsed illumination. It has been discovered that high intensity, pulsed irradiation induces anion remixing to restore original alloy emission energies. Pulsed irradiation, thin film photoremixing appears universal and has been observed across multiple compositions of a given mixed-halide material. For example, photoremixing has been observed in x=0.67, 0.52, and 0.30 FACsPb(IBr)thin films. It has also been observed in other mixed-halide perovskite compositions such as MAPb(IBr)and even in ultrastable, triple cation FAMACsPb(IBr). For high-speed pulsed illumination, necessary fluences are around 1000-10000 uJ/cm, which is on a much smaller order of magnitude than continuous wave intensities. The pulses may cover a duty cycle range of at least 0.1% to 10%, but a range as wide as 0.0001% to 100% may achieve better results.

The frequency range for pulse width modulation is clipped at the low end by the rate constant for photosegregation ˜1 hz. The estimated maximum frequency is 100 khz, but a more practical limit with our drive circuitry is around 100-1000 hz. The maximum frequency is limited by the requirement to have a pulse width significantly greater than the photocarrier lifetime in the material (>10 ns in our case). This is not strictly necessary, but using a pulse width on par with the photocarrier lifetime requires reducing the duty cycle by up to 2 orders of magnitude (and intensity by the same) to achieve the same emission wavelength.

depicts a block diagram of an example of a mixed halide perovskite spectra generator, consistent with some embodiments. In the illustrated example and describe herein, the mixed halide perovskite spectra generatorprovides an example spectra generator. However, other spectra generators may be contemplated and fall under the scope of the present disclosure. It should be noted that the use of “spectra” or “spectrum” herein, unless otherwise noted, should not be limited to an entire range of wavelengths of electromagnetic radiation and may include portions (continuous or non-continuous) of the range of wavelengths of electromagnetic radiation. As discussed above, certain uses cases may only require a specific range or ranges of wavelengths. In some embodiments, the mixed halide perovskite spectra generatormay include a controller, a driver, a photoluminescent device, and a feedback system. The controllermay be electrically or communicatively coupled with the driver. The drivermay be electrically coupled with the photoluminescent deviceto provide a pulsed signal or a continuous signal to the photoluminescent device. The feedback systemmay be in communication with the controllerto provide feedback or to calibrate signals provided by the driverbased on the outputs of the photoluminescent device.

Specifically, in some aspects of the present disclosure, the controllermay include a digital controller such as a microcontroller, a field-programmable gate array (FPGA), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), or other controller that would be apparent to one of skill in the art in possession of the present disclosure. In some embodiments, the controllermay be provided by the processors of the computer systemdescribed in the disclosure describing, discussed below. The type of controller/processor included in controllermay be selected depending on the desired resolution, timing precision, and system requirements.

In some aspects of the present disclosure and discussed in further detail in, the drivermay include a multichannel pulse width modulation (PWM) generator that may be used to fine-tune a duty cycle of a pulsed signal to the photoluminescent deviceand other characteristics of the pulsed signal to cause a light source to provide various intensities of illumination to a mixed halide perovskite film in the photoluminescent device. In other embodiments, the drivermay provide a continuous wave signal that may be adjusted to cause a light source to provide various intensities of illumination to the mixed halide perovskite film in the photoluminescent device. The drivermay be configured to provide a variety of ranges of power to create various intensities of light generated by a light source in the photoluminescent devicesuch that finer wavelength control of the mixed halide perovskite film can be achieved. The pulse width modulation generator may be multichannel to fine tune the signal provided to the photoluminescent deviceor to provide signals to multiple light sources in the photoluminescent deviceor provide signals to multiple photoluminescent devices.

In various aspects of the present disclosure, the mixed halide perovskite spectra generatormay include the photoluminescent device. The photoluminescent devicemay include an emitter that includes a light source and a mixed halide perovskite film that may include one or more emitters of various compositions.illustrate various embodiments of the photoluminescent deviceand emitters and are described in further detail below.

In various aspects of the present disclosure, the mixed halide perovskite spectra generatormay include a feedback system. The feedback systemmay include one or more sensors. In various aspects, the feedback systemmay report the emission wavelength, the intensity, brightness, a measure of brightness within a certain wavelength range, or other characteristics of the light generated by the photoluminescent deviceto the controller. In some embodiments, the feedback systemmay include a spectrometer, a photocell with or without a filter, a photodiode or other sensors or light sensors that would be apparent to one of skill in the art in possession of the present disclosure. The feedback systemmay be used as a calibration tool in which it is used once to characterize the photoluminescent deviceand then the mixed halide perovskite spectra generatormay function without the feedback system. In other aspects, the feedback systemmay be providing real-time feedback to provide real-time adjustments to the signals provided to the photoluminescent deviceor multiple photoluminescent devicesby the driver(s)to achieve a desired photoemission wavelength range/profile (e.g., photoemission peak or peaks having an adjustable center, width, or height) of individual emitters or a photoemission wavelength range/profile of multiple emitters acting together. Because the mixed halide perovskite material changes over time and use, the feedback systemmay be able to communicate “degraded” outputs of the photoluminescent devicesuch that adjustments can be made to the signals provided to the photoluminescent device. As such, the feedback systemmay be used based on a specific use case of the mixed halide perovskite spectra generator.

In some aspects, the mixed halide perovskite spectra generatormay include a heating and cooling system (not illustrated) that is adjacent the photoluminescent deviceand communicatively coupled to the controller. A heater applied to the perovskite active material may accelerate the rate of segregation or mixing. Raising the temperature by 20° C. is enough to reduce the time to reset by a factor of 100. Heating also pushes up the duty cycle range for wavelength tuning facilitating higher intensity operation. Finally, there are some benefits that could be obtained with cooling including sharper emission peaks. The sharper the emission, the finer the degree of control when generating spectra. As such, the feedback systemmay include a temperature sensor such as thermometer to regulate the temperature near the photoluminescent devicesuch that the controllercan regulate the heating and cooling system to generate a temperature at the photoluminescent devicethat produces an photoemissions wavelength range or profile (e.g., photoemission peak or peaks having an adjustable center, width, or height) defined by the controller.